JP5622182B2 - Real-time welding quality judgment device and judgment method - Google Patents

Description

  The present invention relates to a real-time welding quality judgment device and a judgment method, and more particularly to a real-time welding quality judgment device and a judgment method for performing digital signal processing on a measured value of welding energy during welding and comparing it with a reference value.

  In chemical reactions such as chemical reactions, sintering, and vapor deposition, and physicochemical reactions (including processing), various conditions related to reactions such as pressure, temperature, atmosphere, current, and voltage in order to obtain high-quality products. It is important to control properly. For this reason, with the recent development of microprocessors, personal computers, etc., attempts are being made to digitally process and appropriately control the various conditions.

  Also in welding, for example, in arc welding, analog signals for detected currents and voltages are A / D converted and digitized, processed by digital filters, etc., and inverters, and thus currents and voltages are appropriately controlled. (Patent Document 1, Non-Patent Document 1).

  In addition, in resistance welding, an appropriate welding time is set based on the data on the relationship between the welding time and strength, and the result of the sampling inspection for welding performed at the set welding time is reflected to reflect the welding strength. There is also an invention for judging the above in real time (Patent Document 2).

Japanese Patent Laid-Open No. 7-136776 JP 7-136777 A

“Development of digitally controlled arc welding equipment” by Toshiro Uenono, 100th Anniversary Seminar of the Welding Data System Research Committee sponsored by the Japan Welding Society Welding Data System Research Committee (January 14, 2005)

  However, in recent years, user requirements for welding accuracy, efficiency, and cost have become increasingly severe, and the above-described technology is not necessarily sufficient to meet the strict requirements of users.

  The techniques disclosed in Patent Document 1 and Non-Patent Document 1 process the digitized values of the measured voltage and current, and control the inverter and the like, and thus the voltage and current, based on the processing result. However, it is not a technique that directly aims to control welding quality, but to evaluate the effect of short-term changes in welding conditions on welding quality in order to manage welding quality in real time. It is not described from the viewpoint.

  In addition, there are many parameters related to the quality of welding, such as welding speed, shield gas composition, cathode tip angle and arc length if TIG welding, and other non-uniformity depending on the location of the plate material or welding rod to be welded, Simply controlling only the voltage and current is not sufficient to determine the quality of welding.

  Moreover, the technique currently disclosed by patent document 2 manages the intensity | strength of welding only for welding time for resistance welding. However, such management based on welding time alone is difficult to apply to spot welding, etc., and it is also difficult to detect quality defects due to factors other than welding time, such as defects due to electrode tip deterioration. It is.

  Therefore, for quality control of welding at the actual welding construction site, for example, in the case of arc welding, if the current and voltage are within a predetermined range, it is assumed that there is no problem with the construction, and the welding is finished. After that, non-destructive inspection and destructive inspection of the sampled part are performed. For this reason, it is only after the welding has been completed that some defects have been found in the welding. Then, if it is found that a welding defect has occurred, the defect part is reworked at that time, the defective product is screened and repaired, the electrode tip is polished or replaced, etc. This process is necessary, and there are difficulties in terms of welding accuracy, work efficiency, production cost, and the like.

  Furthermore, depending on the welding system and scale, such as an unmanned welding line, once a defect has occurred, if welding continues as it is, it will be necessary to investigate and rework all subsequent welding locations and products. There is. In this case, the maintenance of welding accuracy, work efficiency, production cost, etc. will be greatly affected.

  For this reason, development of a technique capable of determining quality in real time during welding, such as discovery of occurrence of defects in real time, has been desired.

  In order to solve the above problems, the present inventor measured physical quantities related to welding energy such as current and voltage during welding in each processing unit in Japanese Patent Application No. 2009-60771 (Japanese Patent Laid-Open No. 2010-214380). The measured physical quantity is converted into a digital value, signal processing is performed to accurately reflect the quality of the welding, and the result obtained is compared with a reference value to determine the welding quality in the processing unit in real time. A real-time welding quality judgment device was proposed.

That is,
A physical quantity measuring unit for measuring a physical quantity related to the welding energy of the welding part during welding;
The physical quantity measured by the physical quantity measuring unit is converted into a plurality of types and digitized sampling data in real time and in a predetermined procedure for each unit processing time for continuous welding and for each spot for spot welding. Physical quantity digitization department,
A sampling data one-dimensionalization unit that converts each type of digitized sampling data into one-dimensional data in a predetermined procedure including real-time and normalization according to the type and condition of the welding,
A vector expression unit that expresses multidimensional sampling data composed of each type of sampling data, which is made one-dimensional data by the sampling data one-dimensionalization unit, as a one-dimensional sampling vector in a predetermined procedure including real-time and dimensional reduction. When,
A quality determination unit that compares the one-dimensional sampling vector expressed by the vector expression unit with a vector serving as an evaluation reference in real time in a predetermined procedure, and determines the unit processing time or the welding quality of the spot;
We proposed a real-time welding quality judgment device characterized by having

  Specifically, first, physical quantities such as current and voltage related to the welding energy of the welded part are measured during welding, and then the physical quantities are measured every unit processing time if continuous welding, and spot welding if spot welding. For each, a plurality of types and a large number of digitized sampling data are used. Next, each type of digitized sampling data is converted into one-dimensional data by a procedure including normalization according to the type and condition of welding, and further, multidimensional sampling data consisting of each type of one-dimensional data is dimensionally reduced. A one-dimensional sampling vector is obtained by a procedure including Thereafter, the obtained one-dimensional sampling vector is compared with a vector serving as an evaluation criterion to determine the unit processing time or the welding quality of the spot.

  As described above, the proposed real-time welding quality determination apparatus measures and digitizes physical quantities related to welding energy in real time, and includes a procedure including normalization to make it easier to catch changes, variations, or abnormalities. The multi-dimensional data consisting of the one-dimensional data of each physical quantity is converted into a one-dimensional vector by dimensional reduction. For this reason, since a lot of information is finally made into a one-dimensional vector, it is very easy to compare with a preliminarily created evaluation criterion vector.

  The above processing is not limited as long as it is a physical quantity that can be measured, and thus can be widely applied to various types of welding.

  Therefore, it is possible to determine the quality with high accuracy in real time during welding regardless of the type of welding, for example, if a defect occurs during welding, the occurrence of the defect can be easily known in real time. And the quality control of welding is greatly improved, and the accuracy, efficiency and cost of welding are greatly improved.

However, in the above proposal, complex processing such as normalization, multidimensionalization, and dimensional reduction is performed when a sampling vector is created from sampling data. Therefore, the present inventor has completed the present invention as a result of further earnest studies on a technique that can more easily determine the quality of welding in real time without specially performing these complicated processes. It came to do. Hereinafter, first to ninth techniques related to the present invention will be described.

The first technique related to the present invention is:
A physical quantity measuring unit for measuring a physical quantity related to the welding energy of the welding part during resistance welding;
A physical quantity digitizing section that converts the physical quantity measured by the physical quantity measuring section into sampling data digitized in real time by a predetermined procedure;
A sampling data one-dimensionalization unit that converts the digitized sampling data into one-dimensional data in real time in a predetermined procedure according to the welding conditions;
A vector expression unit for expressing the sampling data converted into one-dimensional data by the sampling data one-dimensionalization unit as a one-dimensional sampling vector in real time according to a predetermined procedure;
Obtaining the magnitude of the difference between the one-dimensional sampling vector expressed by the vector expression unit and the vector serving as the evaluation criterion, and based on the ratio of the magnitude of the difference with respect to the standard deviation of the vector serving as the evaluation criterion It is the real-time welding quality determination apparatus characterized by having the quality determination part which determines the quality of this.

This technology also basically uses the digital signal processing to measure and extract the physical quantities that have a significant effect on welding, such as current and voltage, and to characterize and extract the quality of the welds from the measured values, as in the proposed invention. And the quality of the weld is determined by evaluating the processed signal.

  Specifically, first, a physical quantity such as a current and a voltage related to the welding energy of a welding part is measured during resistance welding such as spot welding, seam welding, and butt welding, and then the physical quantity is measured for each welding, for example, In the case of spot welding, sampling data digitized in real time is used for each spot. Next, the digitized sampling data is converted into one-dimensional data in real time according to a predetermined procedure according to the welding conditions, and the sampling data converted into the one-dimensional data is converted into one-dimensional data in real time according to a predetermined procedure. Sampling vector. Then, the magnitude of the difference between the obtained one-dimensional sampling vector and the evaluation reference vector is obtained, and the quality of the welding is determined based on the ratio of the difference magnitude to the standard deviation of the evaluation reference vector. To do.

As described above, the real-time welding quality determination apparatus according to the present technology measures and digitizes the physical quantity related to the welding energy in real time, performs one-dimensional data according to a predetermined procedure, and further performs processing to obtain a one-dimensional vector. ing. Since this one-dimensional vector includes a lot of information on the welding quality, the magnitude of the difference obtained from the previously created evaluation standard vector is the vector of the evaluation standard. By determining the ratio to the standard deviation, the quality of the weld can be determined very easily in real time. As a result, welding quality management is greatly improved, and welding accuracy, efficiency, and cost management are greatly improved.

In the present technology , when determining the quality, the magnitude (absolute value) of the difference between the one-dimensional sampling vector and the vector serving as the evaluation criterion is obtained, and the standard deviation of the vector serving as the evaluation criterion for the magnitude of the difference is obtained. Since the quality is determined based on the ratio, it is not necessary to perform complicated processing such as normalization, multidimensionalization, and dimensional reduction as in the proposed invention.

That is, the process for obtaining the ratio of the magnitude of the difference between the one-dimensional sampling vector and the evaluation reference vector with respect to the standard deviation of the evaluation reference vector is a process for obtaining a deviation from the weld quality evaluation reference in the welding. It is a process that brings about a result substantially equivalent to the normalization described above. For this reason, in the present technology , the normalization process is not performed in the intermediate processing stage, and the accurate quality is obtained by performing a process similar to the normalization using a simple method in the final quality determination process stage. Judgment can be made.

  Similarly, since it is not necessary to specially provide complicated processing steps such as multi-dimensionalization and dimensional reduction, the entire processing can be simplified, and the welding quality can be determined more easily.

The details will be described later in “Mode for Carrying Out the Invention”, but main terms used in the present technology will be described in advance.

First, “real-time” as used in this technology includes cases that are considered to be real-time in practical use. For convenience of other processing in the computer, for example, checking of other welding locations when a large number of weldings are performed simultaneously. This includes cases where there is a delay of about 2 or 3 seconds due to restrictions determined by the length of the unit processing time, and other restrictions determined by the convenience of other machinery and processes in the entire factory and welding equipment. .

  Next, examples of the “physical quantity related to welding energy” and “one-dimensional data” include combinations of interelectrode voltage and current (physical quantity) and resistance value (one-dimensional data) described later. In addition, combinations of changes in electrode pressure (physical quantities) and resistance values (one-dimensional data), combinations of current and voltage (physical quantities), and heat quantities (one-dimensional data) can also be mentioned. Also, electrode displacement (movement amount), resistance gradient, and the like can be employed as physical quantities.

  Next, “physical quantity measurement” is usually performed continuously and in analog form, and an analog value is digitized by a physical quantity digitizing unit. If possible, the physical quantity may be acquired directly as a digital value.

  Secondly, “digitized sampling data” is not just data with measured values as digital values, but also takes absolute values of measured values that have plus or minus, such as AC power supply voltage, Also included is data that has been subjected to processing for removing.

  Next, “the sampling data is converted into one-dimensional data by the one-dimensionalization unit” means that a plurality of types of digitized sampling data for a plurality of types of physical quantities created by the physical quantity digitizing unit are This means that the physical quantity is one-dimensional.

  Next, “represent as a one-dimensional sampling vector” means that each sampling point indicated by multi-dimensional sampling data is indicated by the vector expression unit as one type of numerical value indicating the quality of the welding ( It is expressed).

  The “difference magnitude” indicates the absolute value of the difference between the one-dimensional sampling vector and the evaluation reference vector.

In addition, the real-time welding quality determination apparatus of the present technology is also provided with a recording unit, an input unit, and the like for predetermined determination criteria.

The second technique related to the present invention is:
The quality judgment unit
When the one-dimensional sampling vector is Rt, the evaluation standard vector is Rs, and the standard deviation is Sp,
Each of Rt, Rs, and Sp is divided into n in time series from 1 to n,
Real-time welding quality according to the first technique , wherein the quality of the welding is determined by comparing the total sum D of D _n obtained by the following formula with a predetermined allowable reference value. It is a determination device.
D _n = | (Rt _n −Rs _n ) | / Sp _n
(However, Rt _n : n-th data of the one-dimensional sampling vector Rt
Rs _n : n-th data of vector Rs serving as the evaluation criterion
Sp _n : nth data of standard deviation Sp)

The present technology specifically defines the processing in the quality determination unit described in the first technology , and includes a one-dimensional sampling vector Rt divided into n in time series and a vector Rs serving as an evaluation criterion. and based on the respective n-th data of the standard deviation Sp, seeking n-th data D _n of the D indicating the degree of deviation from the criteria.

For this reason, the sum D of the n data from D ₁ to D _n is the sum of the deviations from the evaluation standard in the welding, and the larger this D (welding quality distance), the worse the welding quality. If it exceeds a predetermined allowable reference value, it is determined as defective.

The vector Rs serving as an evaluation criterion is obtained by sampling a model hitting point in advance and forming a vector from the data. This exemplary hit point need not be a single hit point, but may be a vector formed from data at a plurality of hit points. As the number of hit points increases, the accuracy of determination increases. However, this time, since Rs _n is the n-th data in each weld point is created on average (simple average), also Sp _n is created as the standard deviation of the n-th data in each RBI, the processing It takes time. For this reason, the vector Rs serving as an evaluation criterion is usually created from data of about 3 to 4 dots as a practically unaffected level.

  The “predetermined allowable reference value” can be set as appropriate based on, for example, previous experiments and experiences.

The third technique related to the present invention is:
The quality judgment unit
The n-th data Rt _n of the one-dimensional sampling vector Rt is subjected to a predetermined classification according to the elapsed time from the start of welding,
By the total sum D of the weighted D _n is obtained by relative D _n multiplied by the weighting coefficients corresponding to the classification, compared with the allowable reference value is determined in advance, determining the quality of the welding The real-time welding quality determination device according to the second technique , characterized in that:

  When performing resistance welding, for example, a large resistance value (initial resistance) is indicated by the surface roughness of the base material in the initial stage of energization, and the surface roughness is eliminated and stable after a certain degree of welding. The resistance value is shown. There is also a variation in resistance value due to fluctuations in power supply during welding.

  For this reason, when determining the quality of welding, it is necessary to perform determination in consideration of these variations.

In this technique , since the welding quality is determined after weighting according to the elapsed time from the start of welding to each of the created one-dimensional sampling vectors, the welding quality can be determined more accurately. Can do.

  In addition, about specific weighting, time passage from the start of welding is divided into parts where different situations characteristically appear, for example, an initial part, an intermediate part, and a late part, and weighting is appropriately performed corresponding to each part. Is done.

In addition, as the “predetermined allowable reference value” in the present technology , an allowable reference value obtained by performing the same weighting as the weighting performed on the D _n is employed.

The fourth technique related to the present invention is:
A new electrode recognition unit that recognizes that the electrode of the resistance welder has been replaced,
Upon receiving notification that the electrode has been newly exchanged from the new electrode recognition unit, the physical quantity measurement unit, the physical quantity digitization unit, the sampling data one-dimensionalization unit, and the vector expression unit are processed in accordance with the notification. Creates a one-dimensional sampling vector for each predetermined number of welds from the current and voltage at the predetermined number of times after the welding, and then creates a vector as an evaluation criterion based on the created one-dimensional sampling vectors. An evaluation reference vector creation unit for
The quality judgment unit
Upon determination of the quality of the welding of the predetermined number of times later, real-time welding according to any one of the first technique to a third technique, which comprises using the vector in which the evaluation reference vector creation part becomes evaluators created It is a quality judgment device.

  In resistance welding in large-scale factories such as automobile manufacturing factories, there are a wide variety of conditions such as the thickness, type and combination of base materials, welding specifications, and electrode and welding machine specifications. For this reason, it is practically difficult to create a vector as an evaluation criterion in advance for all of these conditions.

In the present technology , a predetermined number of one-dimensional sampling vectors are created from welding performed immediately after the electrodes are replaced, and a vector serving as an evaluation criterion is created based on the one-dimensional sampling vectors.

  This is because it is known that the quality of welding immediately after replacing the electrodes is good, and there is no problem in adopting these one-dimensional sampling vectors as an example for creating an evaluation standard.

The fifth technique related to the present invention is:
The physical quantity measuring unit measures voltage and current,
The physical quantity digitizing unit uses the measured voltage and current as sampling data digitized in real time for each welding,
Said sampling data one-dimensional section, the two kinds of sampling data for the digitized voltage and current, to no first technique, characterized by a one-dimensional data regarding the resistance in real time with a predetermined procedure 4 A real-time welding quality determination device according to any one of the above techniques .

In the present technology , the interelectrode voltage V and the current I are adopted as specific physical quantities, and based on the obtained values, a one-dimensional sampling vector for the resistance R calculated by the equation R = V / I. Have created.

  This is because the heat generation principle in resistance welding can be explained by the electrical resistance at the time of welding, and the role of electrical resistance in resistance welding is very large, and the occurrence of poor welding quality is reflected in the electrical resistance value. It is.

The sixth technique related to the present invention is:
A physical quantity measuring unit for measuring a temporal change in the voltage between the electrodes of the welded spot as a physical quantity during spot welding;
A physical quantity digitizing unit that converts the temporal change of the interelectrode voltage measured by the physical quantity measuring unit into sampling data digitized in real time by a predetermined procedure;
Using the digitized sampling data, a Mahalanobis distance calculation unit that calculates the Mahalanobis distance in real time in a predetermined procedure;
Based on the control limit shown in the following formula created from Mahalanobis distance data in spot welding, which is a separately calculated evaluation standard, as a threshold, the Mahalanobis distance data calculated by the Mahalanobis distance calculation unit is sorted and included below the threshold A real-time welding quality judgment device comprising a quality judgment unit for judging the quality of the welding based on the data amount of the Mahalanobis distance.
μ + 3σ
μ: Average value of Mahalanobis distance data in evaluation spot welding σ: Standard deviation of Mahalanobis distance data in evaluation spot welding

  In statistical processing using a characteristic value having a correlation as a parameter, a method of evaluating using the Mahalanobis distance is known. This evaluation based on the Mahalanobis distance is obtained by evaluating the distance between points arranged in a two-dimensional space in a metric space in consideration of the correlation.

Specifically, first, as in the first technique , a physical quantity at one hit point during welding is measured every measurement unit time, digitized in real time to obtain sampling data, and then the Mahalanobis distance is set for each sampling data. calculate. Separately, for welding as an evaluation standard, the Mahalanobis distance is similarly calculated (reference data) to obtain an average value μ and a standard deviation σ. Thereafter, the 3σ limit, which is a reasonable control limit in various statistical methods, that is, μ + 3σ is set as a threshold value (since the Mahalanobis distance is 0 or more, only μ + 3σ is used), each of the above obtained Sort the Mahalanobis distance of sampling data.

  However, for example, when voltage and current are measured as physical quantities and the quality is determined using the Mahalanobis distance, the welding quality can be appropriately determined from the sampling data of the Mahalanobis distance in the case of arc welding or TIG welding. However, in the case of spot welding, it was found that electrode consumption and nugget formation cannot be determined at the obtained Mahalanobis distance, and the welding quality cannot be sufficiently determined.

  In view of this, the present inventor has studied a new physical quantity capable of appropriately performing quality determination using the Mahalanobis distance in spot welding. As a result, the temporal change of the interelectrode voltage is more directly related to the welding quality. The temporal change of the interelectrode voltage is measured as a physical quantity, and the measured temporal change of the interelectrode voltage is measured. It has been found that by performing appropriate processing using the Mahalanobis distance, it is possible to determine electrode consumption and nugget formation status, and to appropriately determine welding quality.

The seventh technique related to the present invention is:
In the sixth technique , the quality determination unit obtains the welding quality by the following formula based on the data amount Nu of the Mahalanobis distance and the total data amount Na included below the threshold, and performs quality determination. It is a real-time welding quality judging device of statement.
Weld quality (%) = (Nu / Na) × 100

  As described above, if the data amount (number) of the Mahalanobis distance included below the threshold is large, it can be determined that good quality welding is being performed, so by defining the ratio to the total data amount as welding quality The quality can be determined quantitatively.

The eighth technique related to the present invention is:
The physical quantity digitizing unit uses the temporal change of the interelectrode voltage measured by the physical quantity measuring unit as two-dimensional sampling data of the interelectrode voltage and time digitized in real time by a predetermined procedure,
The Mahalanobis distance calculation unit draws straight lines parallel to the time axis at equal intervals with respect to the two-dimensional sampling data of the interelectrode voltage and time digitized by the physical quantity digitizing unit, The number of intersections with the straight line is obtained as differential data, and the sum of the data amount of the voltage between the electrodes in the area above each straight line is obtained as integral data, and the differential data and the integral data are used in real time. The real-time welding quality determination apparatus according to the sixth technique or the seventh technique , wherein the Mahalanobis distance is calculated.

  As described above, the temporal change of the interelectrode voltage is more directly related to the welding quality, and the two-dimensional data of the interelectrode voltage and the time created based on the temporal change of the measured interelectrode voltage. The above-mentioned differential data and integral data obtained from (electrode voltage waveform) contain information on the frequency and amplitude of the waveform that affects the welding quality, and information on their distribution. By comparing the Mahalanobis distance data calculated by using the Mahalanobis distance data, which is the evaluation standard, it is possible to determine the electrode wear and nugget formation status, and to determine the welding quality appropriately in real time at high speed. be able to.

  That is, the data including information that influences the welding quality is arranged in the differential-integral secondary space in consideration of the above-described correlation, and the Mahalanobis distance is calculated, and the result is calculated as μ + 3σ as described above. By allocating the threshold value, it is possible to appropriately determine the quality of welding in real time at high speed.

  In the above, the number of intersections between the two-dimensional data and each straight line is expressed as differential data. The intersection with each straight line can be interpreted as an intersection with the time axis and has a differential characteristic. Because it can be regarded as data. Similarly, the sum of the data amount of the interelectrode voltage in the area above each straight line is expressed as integral data. The sum of the data amount of the interelectrode voltage can be interpreted as an area, and is integral with time. This is because it can be regarded as data having characteristics.

  The number of “straight lines parallel to the time axis” is generally preferably about 500 to 1500.

  In addition, Mahalanobis distance data, which serves as an evaluation standard, may be created in advance by appropriately sampling model points, but spot welding is an automatic process in which a series of welding processes are automatically performed by simply pressing the foot switch. Since it is welding, there is almost no difference in the judgment result of the welding quality even if the number of hit points is increased.

The ninth technique related to the present invention is:
A real-time welding quality judgment method characterized in that the quality of welding is judged in real time using the real-time welding quality judgment device according to any one of the first to eighth techniques .

In the present technology , the technology described in any of the first to eighth technologies, which is an invention of the device, is regarded as a method technology .

  The present invention has been made on the basis of the above-described technologies, and the invention according to claim 1
  A physical quantity measuring unit for measuring a physical quantity related to the welding energy of the welding part during resistance welding;
  A physical quantity digitizing section that converts the physical quantity measured by the physical quantity measuring section into sampling data digitized in real time by a predetermined procedure;
  A sampling data one-dimensionalization unit that converts the digitized sampling data into one-dimensional data in real time in a predetermined procedure according to the welding conditions;
  A vector expression unit for expressing the sampling data converted into one-dimensional data by the sampling data one-dimensionalization unit as a one-dimensional sampling vector in real time according to a predetermined procedure;
  Obtaining the magnitude of the difference between the one-dimensional sampling vector expressed by the vector expression unit and the vector serving as the evaluation criterion, and based on the ratio of the magnitude of the difference with respect to the standard deviation of the vector serving as the evaluation criterion A quality judgment unit that judges the quality of
Have
  The quality judgment unit
  When the one-dimensional sampling vector is Rt, the evaluation standard vector is Rs, and the standard deviation is Sp,
  Each of Rt, Rs, and Sp is divided into n in time series from 1 to n,
  D obtained by the following formula _n The quality of the welding is determined by comparing the sum D of the above with a predetermined allowable reference value.
This is a real-time welding quality judgment device.
      D _n = | (Rt _n -Rs _n ) | / Sp _n
      (However, Rt _n : Nth data of one-dimensional sampling vector Rt
              Rs _n : Nth data of the vector Rs serving as the evaluation criterion
              Sp _n : Nth data of standard deviation Sp)

  The invention according to claim 2
  The quality judgment unit
  N-th data Rt of the one-dimensional sampling vector Rt _n For a specific classification according to the elapsed time from the start of welding,
  D _n Weighted D obtained by multiplying the weighting coefficient corresponding to the classification by _n The quality of the welding is determined by comparing the sum D of the above with a predetermined allowable reference value.
The real-time welding quality determination apparatus according to claim 1, wherein:

  The invention according to claim 3
  A new electrode recognition unit that recognizes that the electrode of the resistance welder has been replaced,
  Upon receiving notification that the electrode has been newly exchanged from the new electrode recognition unit, the physical quantity measurement unit, the physical quantity digitization unit, the sampling data one-dimensionalization unit, and the vector expression unit are processed in accordance with the notification. Creates a one-dimensional sampling vector for each predetermined number of welds from the current and voltage at the predetermined number of times after the welding, and then creates a vector as an evaluation criterion based on the created one-dimensional sampling vectors. An evaluation reference vector creation unit
Have
  The quality judgment unit
  When determining the quality of welding after the predetermined number of times, a vector serving as an evaluation criterion created by the evaluation criterion vector creating unit is used.
The real-time welding quality determination apparatus according to claim 1 or 2, wherein the apparatus is a real-time welding quality determination apparatus.

  The invention according to claim 4
  The physical quantity measuring unit measures voltage and current,
  The physical quantity digitizing unit uses the measured voltage and current as sampling data digitized in real time for each welding,
  The sampling data one-dimensionalization unit converts the two types of sampling data regarding the digitized voltage and current into one-dimensional data related to resistance in real time according to a predetermined procedure.
The real-time welding quality judgment device according to any one of claims 1 to 3, wherein

  The invention according to claim 5
  A physical quantity measuring unit for measuring a temporal change in the voltage between the electrodes of the welded spot as a physical quantity during spot welding;
  A physical quantity digitizing unit that converts the temporal change of the interelectrode voltage measured by the physical quantity measuring unit into sampling data digitized in real time by a predetermined procedure;
  Using the digitized sampling data, a Mahalanobis distance calculation unit that calculates the Mahalanobis distance in real time in a predetermined procedure;
  Based on the control limit shown in the following formula created from Mahalanobis distance data in spot welding, which is a separately calculated evaluation standard, as a threshold, the Mahalanobis distance data calculated by the Mahalanobis distance calculation unit is sorted and included below the threshold A quality determination unit for determining the quality of the welding based on the data amount of the Mahalanobis distance
Have
  The physical quantity digitizing unit uses the temporal change of the interelectrode voltage measured by the physical quantity measuring unit as two-dimensional sampling data of the interelectrode voltage and time digitized in real time by a predetermined procedure,
  The Mahalanobis distance calculation unit draws straight lines parallel to the time axis at equal intervals with respect to the two-dimensional sampling data of the interelectrode voltage and time digitized by the physical quantity digitizing unit, The number of intersections with the straight line is obtained as differential data, and the sum of the data amount of the voltage between the electrodes in the area above each straight line is obtained as integral data, and the differential data and the integral data are used in real time. Calculate Mahalanobis distance
This is a real-time welding quality judgment device.
    μ + 3σ
        μ: Average value of Mahalanobis distance data in spot welding as the evaluation standard
        σ: Standard deviation of Mahalanobis distance data in spot welding as an evaluation standard

  The invention according to claim 6
  The quality determination unit performs quality determination by obtaining the welding quality by the following formula based on the data amount Nu of the Mahalanobis distance and the total data amount Na included below the threshold.
The real-time welding quality determination apparatus according to claim 5, wherein:
      Weld quality (%) = (Nu / Na) × 100

  The invention described in claim 7
  A real-time welding quality judgment method characterized in that the quality of welding is judged in real time using the real-time welding quality judgment device according to any one of claims 1 to 6.

  In the present invention, if a defect occurs during welding, a technique capable of easily determining the quality in real time during welding, such as finding the occurrence of a defect in real time, can be provided. Quality control is greatly improved, and welding accuracy, work efficiency, production cost, etc. are greatly improved.

It is a figure which shows the outline of the process in the 1st Embodiment of this invention. In the 1st Embodiment of this invention, it is a figure which shows the process of preparing vector Rs used as evaluation criteria. It is a figure which shows the process of calculating | requiring a welding quality distance in the 1st Embodiment of this invention. In the 1st Embodiment of this invention, it is a figure which shows the relationship between the fracture inspection result of a welding location, and the corresponding welding quality distance. It is a figure which shows typically transition of the flow of the electricity at the time of spot welding. It is a figure which shows a mode that a nugget is formed in spot welding. It is a figure which shows that the time change of the resistance value in spot welding and the change itself differ with cases. It is a figure which shows the relationship between the welding inspection distance and the corresponding welding quality distance in the 3rd Embodiment of this invention. It is a figure which shows notionally the structure of the spot welder and its quality determination apparatus which were used in the 4th Embodiment of this invention. It is a figure which shows the measurement result of the average voltage in the 4th Embodiment of this invention, and the result in a twist fracture test. It is a figure which shows the result of having calculated | required welding quality by making a voltage and an electric current into parameters in the 4th Embodiment of this invention. It is a figure which shows the result of having calculated | required distribution of Mahalanobis distance by using a voltage and an electric current as a parameter in the 4th Embodiment of this invention. It is a figure which shows notionally the time change of the voltage between electrodes in the spot welding of a mild steel material. In the 4th Embodiment of this invention, it is a figure which shows notionally the temporal change of the voltage between electrodes in the 51st hit point and the 451th hit point. In the 4th Embodiment of this invention, it is a figure explaining the determination method of the welding quality which uses the time change of the voltage between electrodes as a parameter. In the 4th Embodiment of this invention, it is a figure which shows the result of having calculated | required the distribution of the Mahalanobis distance in the 51st hitting point and the 451st hitting point using the temporal change of the voltage between electrodes as a parameter. In the 4th Embodiment of this invention, it is a figure which shows the result of having calculated | required welding quality based on the Mahalanobis distance using the time change of the voltage between electrodes.

  Hereinafter, the present invention will be described based on embodiments. Note that the present invention is not limited to the following embodiments. Various modifications can be made to the following embodiments within the same and equivalent scope as the present invention.

  In the first to third embodiments shown below, in spot welding using resistance welding performed in an automobile production line or the like, two types of current and voltage were measured as a plurality of physical quantities, and measured 2 A one-dimensional sampling vector (resistance) is generated by processing data of physical quantities of seeds in real time, and the generated vector and a vector serving as an evaluation criterion are calculated using a predetermined deviation including a standard deviation of the vector serving as an evaluation criterion. The comparison is made by the procedure, and the real-time quality judgment is performed.

(First embodiment)
FIG. 1 shows an outline of processing in the first embodiment. In FIG. 1, 10 is a physical quantity measurement unit, 20 is a physical quantity digitization unit, 30 is a sampling data one-dimensionalization unit and vector expression unit, and 40 is a quality judgment unit.

1.1 Creation of Dimensional Sampling Vector First, the physical quantity measuring unit 10 measures current and voltage as analog data at a predetermined time series interval for one spot during spot welding, and then the physical quantity digitizing unit 20 The continuous data of the current A and the voltage V measured by the physical quantity measuring unit 10 is digitized.

  Next, in the sampling data one-dimensionalization unit and the vector expression unit 30, the resistance R is calculated using the digitized current A and voltage V by the equation R = V / A, and one-dimensional of each time series. After obtaining the sampling data, a one-dimensional sampling vector Rt is created for the resistance at one spot during spot welding.

2. Determination of Welding Quality Next, in the quality determination unit 40, the magnitude of the difference (Rt−Rs) between the created one-dimensional sampling vector Rt and the vector Rs (standard deviation Sp) that is a previously created evaluation criterion. A ratio D (welding quality distance) of | (Rt−Rs) | to Sp is obtained, and the magnitude is compared with a predetermined allowable value to determine the welding quality.

  FIG. 2 is a diagram illustrating a process of preparing a vector Rs serving as an evaluation criterion. In FIG. 2, four hit points are selected as model hit points, and vectors Rs (1) to Rs (4) corresponding to the hit points are obtained. Each vector is divided in time series into 1, 2,..., N,.

Then, the standard deviation Sp _{is, Rs (1) n ~Rs (} 4) n obtained from each of the n-th data standard deviation Sp _n, can be represented by words Sp = Sp _n. Further, Rs _{is, Rs (1) n ~Rs (} 4) n each n-th Rs _n which data are simply averaged, that is, represented by Rs = Rs _n.

  Using Rs and Sp prepared as described above and Rt divided into n pieces in time series from 1 to n (see FIG. 3), the welding quality is determined by the following procedure.

First, based on the nth data of each of Rt, Rs, and Sp, the magnitude Dn of the _nth data of the welding quality distance D is obtained using the following equation.
D _n = | (Rt _n −Rs _n ) | / Sp _n

The total sum of D _n obtained is the welding quality distance D (D = ΣD _n ), and the quality of welding is determined by comparing with a predetermined allowable reference value.

  As described above, since the above-described process is a process that produces a result substantially equivalent to normalization, in the present embodiment, it is not necessary to perform the normalization process at an intermediate process stage. In addition, since complicated processing such as multi-dimensionalization and dimensional reduction is not required, it is possible to make an accurate determination of welding quality while being simpler.

  FIG. 4 shows a result of performing a break inspection of the welded portion after performing a large number of spot weldings and examining a relationship with a corresponding welding quality distance D. The welding conditions at this time were dome-shaped φ6R40 electrodes, energized for 5 seconds per minute with a direct current power source of 8.5 kA, that is, a welding time of 1/12 seconds. As the base material, a JAC270D steel plate (thickness 0.7 mm) was used.

  In FIG. 4, the left vertical axis is the fracture diameter (mm), the right vertical axis is the welding quality distance D, and the horizontal axis is the number of striking points (the order of spot welding points arranged in time series). The line consisting of □, ●, and ■ shown at the top indicates the fracture diameter. Among these, □ indicates that the base metal (base metal) of the welded portion was broken (B fracture). The black circle indicates that the welded portion has undergone shear fracture (S fracture), and the solid square indicates a ring nugget (the center portion of the nugget is not melted).

  Also, the line shown at the bottom shows the weld quality distance.

  From FIG. 4, in this case, it can be seen that the welding quality distance varies greatly from around the point where the number of hit points slightly exceeds 7000, and the turbulence occurs. At this point, the welding operation is stopped and the electrode is replaced. It is understood that it is necessary to deal with.

  In this way, it is possible to accurately check the welding quality with an appropriate standard in real time, so that it does not rely on the individual skill of timing the electrode replacement by hitting with a chisel as in the past, in real time. Thus, the timing of electrode replacement can be achieved simply and reliably.

  The data obtained during the real-time welding quality judgment using the real-time welding quality judgment device of the present invention or the data newly obtained in actual use is stored as a database. You may make it utilize for the determination of the quality of subsequent welding in real time.

(Second Embodiment)
The second embodiment is different from the first embodiment in that weighting is performed on the time from the start of welding when evaluating the welding quality.

  Points of interest in the second embodiment will be described with reference to FIGS.

  FIG. 5 is a diagram schematically showing the transition of the flow of electricity during spot welding, in which the upper stage is a diagram showing the outline of the configuration of spot welding, 80 is an electrode, 90 is a base material, and 98 Indicates a spot weld. The lower part of FIG. 5 is an enlarged view of the spot welded portion 98 and shows the flow of electricity at the time of spot welding. The left figure shows the state when the spot welding is started, and the right figure shows the state when the steady state is reached. 99 is a cavity indicating the roughness of the surface of the base material 90.

  As shown in the lower left diagram of FIG. 5, when spot welding is started, the surface of the upper and lower base materials 90 remains rough, and there is a cavity 99, so that the mutual contact becomes uneven, The flow is greatly disturbed and shows a large resistance value. On the other hand, in the steady state, as shown in the lower right diagram, since the mutual contact becomes uniform, the flow of electricity becomes stable and shows a small resistance value.

  At this time, as shown in FIG. 6, the base material 90 is melted in the order from the portions 95 to 96 and 97 sandwiched between the upper and lower electrodes 80, and solidifies with the end of energization of the portions to form the nugget. It is formed.

  FIG. 7 is a diagram illustrating a temporal change in resistance value in spot welding and a difference itself depending on the case. In FIG. 7, the vertical axis represents resistance, and the horizontal axis represents elapsed time. From FIG. 7, it can be seen that the resistance at the start of spot welding is extremely large, and the variation depending on the case is relatively large, and becomes substantially constant after entering a steady state. In addition, even after entering the steady state, it can be seen that there is a change near the end in some cases.

  For this reason, when determining the quality of spot welding, the data collected evenly over the entire welding is not used for the evaluation as it is, for example, the group immediately after the start of welding, the first group in the subsequent steady state, More accurate determination can be obtained by dividing the entire weld into groups according to the elapsed time, such as dividing into three groups in the latter half, and further weighting the data of each group.

(Third embodiment)
The third embodiment relates to a mode in which a real-time welding quality determination apparatus itself creates a vector that serves as an evaluation criterion.

  Many of the causes of quality defects in spot welding are electrode wear, so there is little probability that welding immediately after electrode replacement will be defective.

  Also, at the start of welding immediately after electrode replacement, the operator shall check whether the replacement has been made properly, and also monitor the operation of the entire welder after replacement, and immediately check the welding equipment if an abnormality is found. It is usually performed to stop and perform necessary maintenance. Also from this aspect, there is little probability that the welding immediately after the electrode replacement will be defective.

  Therefore, in the third embodiment, the quality of the first three spot welding is assumed to be good, and the evaluation standard is obtained by arithmetically averaging the one-dimensional sampling vectors expressed by the vector expression unit for the three positions. And is used to evaluate the quality of subsequent spot welding.

  In the real-time welding quality determination apparatus of the third embodiment, a new electrode recognition unit that recognizes electrode replacement and an evaluation standard based on a one-dimensional sampling vector in the first three spot weldings And an evaluation reference vector creation unit for creating a vector.

  When the new electrode recognition unit recognizes the replacement of the electrodes, the evaluation reference vector creation unit adjusts the physical quantity measurement unit, the physical quantity digitization unit, the sampling data one-dimensionalization unit, and the vector expression unit for the first three spot weldings. Processing is performed to create a new evaluation reference vector.

  After creating a new evaluation standard vector, the welding quality is determined according to this evaluation standard vector.

  FIG. 8 shows the result of performing a break inspection of the welded part and examining the relationship with the corresponding weld quality distance based on the third embodiment after performing many spot welds. Welding conditions at this time were performed using a dome-shaped φ6R40 electrode with an AC power supply of current 6.2 kA and a power supply time of 13 cycles @ 60 Hz and a pressure of 150 kgf.

  As can be seen from FIG. 8, in the case of this example, the welding quality distance is small when the number of hit points is 200 or less, and the weld quality distance increases rapidly from the vicinity where the number of hit points exceeds 1600.

(Fourth embodiment)
The fourth embodiment relates to determining the quality of spot welding using the Mahalanobis distance. In the following, as a specific experimental example of spot welding, two-layer lap welding of GA material will be described.

  First, it will be described that the welding quality cannot be sufficiently determined when the quality determination is performed using the Mahalanobis distance with the voltage and current as physical quantities.

  FIG. 9 conceptually shows the configuration of the spot welder and its quality judgment device used in this experiment. In FIG. 9, 1 is a direct current spot welding machine with a servo-type pressurizing method of the power source, 2 is a toroidal coil, 3 is a weld monitor for measuring the voltage between electrodes and welding current (Miyachi Technos, MG3), 4 is This is a PC (Personal Computer) that performs processing based on the interelectrode voltage and welding current measured in the weld monitor 3. In addition, the measurement of the voltage between electrodes by the weld monitor 3 was performed by using both electrode tips sandwiching the material to be welded, and the measurement of the welding current was performed by using the toroidal coil 2 at a sampling interval of 0.38 ms.

  Then, as the GA material, two 0.7 mm thick JAC270D materials were used, and spot welding was performed with a pressing force of 2.5 kN. At this time, the shape of the tip of the electrode tip was DR type (6 mmφ, 40R), the energization time of each spot was 25 cycles, and the energization current was 10.0 kA. In addition, the sampling frequency was set to 2 points for every 50 points.

  Separately, a torsional fracture test of the welded member was performed, the fracture diameter (mm) was measured, and the fracture status was confirmed. Note that the fracture diameter was measured at two locations of the minor axis and the major axis with respect to the welded portion, and the average value was taken as the fracture diameter.

  In FIG. 10, the measurement result of the average voltage in spot welding and the result in a twist fracture test are shown. In FIG. 10, the abscissa indicates the striking point (the number of striking points counted from the first spot welding), the ◯ and ● marks indicate the fracture diameter (mm), and the ■ mark indicates the average voltage (V). In addition, as for the breaking condition, a circle indicates a fracture of the base material, and a circle indicates a fracture at the welded portion.

  From FIG. 10, it can be seen that in the vicinity of 400 to 900 striking points, a sufficient nugget is not formed, the fracture diameter decreases, and the fracture condition changes from fracture of the base material to fracture of the welded portion. This is presumably because a part of the electrode tip softens and changes in the vicinity of 400 hit points and the electrode tip is deformed from a circular shape to a ring shape.

  In addition, after the 900 striking points, it can be seen that the formation of the nugget is recovered, the fracture diameter is increased, and the fracture condition is the fracture of the base material, so that good welding is obtained. The reason why good welding was obtained after 900 points in this way is considered to be that the shape of the electrode tip was re-formed by repeating welding, and the state of the electrode was somewhat recovered, so that nuggets were easily formed. .

  Next, the Mahalanobis distance is obtained using the measured physical quantities (voltage and current) as parameters in the same manner as in the first embodiment, and the threshold value μ + 3σ created from the average value μ and the standard deviation σ of reference data serving as an evaluation standard. From the comparison, the welding quality in the welding was calculated. The results are shown in FIG. As shown in FIG. 11, when voltage and current are used as parameters, although it is slightly decreased near the 300 hit points, the other hit points have moved around 98%. What is the change in fracture diameter shown in FIG. 10? Not supported. From this result, it can be seen that when the parameters are voltage and current, the difference in nugget formation cannot be determined using the Mahalanobis distance, and the quality of the weld at the welded part cannot be determined sufficiently.

  This is because the nugget formation state is clearly different, for example, the 51st spot (an arrow on the left side of FIG. 10) where a good fracture diameter was obtained, and the 451th spot (FIG. 10) where the fracture diameter decreased due to electrode consumption. This is because, when the Mahalanobis distance in the right arrow) is viewed, there is almost no difference between the 51st spot in the upper row (1) and the 451 spot in the lower row (2), as shown in FIG.

  Next, it will be described that the welding quality can be appropriately determined when the quality determination is performed using the Mahalanobis distance with the temporal change of the interelectrode voltage as a physical quantity.

  As described above, the inventor of the present invention paid attention to the temporal change of the interelectrode voltage that is more directly related to the welding quality as the physical quantity instead of the voltage and the current. FIG. 13 is a diagram conceptually showing a temporal change in the interelectrode voltage measured in spot welding of the mild steel plate such as JAC270D described above, the vertical axis is the interelectrode voltage (V), and the horizontal axis is the time. (Ms).

  In FIG. 13, between AB, it has shown that contact resistance increased by contact resistance heat_generation | fever. And between BC, it has shown that the minute convex part of the contact surface melt | dissolved by contact resistance heat_generation | fever, and contact resistance disappeared. In addition, although the resistance increases due to a temperature rise due to resistance heat generation at the energized portion between the CDs, it shows that nuggets are starting to be formed. When the nugget starts to be formed, the current path becomes wide, while the temperature rise of the welded portion is saturated, so that the resistance starts decreasing from the point D and the voltage between the electrodes starts to decrease. Nuggets expand rapidly between DEs and saturate nugget growth between FGs. After the point G, the current density is reduced due to the expansion of the corona bond area, and the temperature of the welded part is rather lowered despite the energization.

  And the change with time (voltage waveform between electrodes) of the above-mentioned interelectrode voltage is, for example, as shown in FIG. 14, at the 51st point (as described above, a good fracture diameter is obtained and the welding quality is good). Since there is a clear difference in the voltage waveform between the electrodes at the 451st point (as described above, the weld diameter is reduced and the welding quality is poor), the temporal change in the voltage between the electrodes is measured as a physical quantity. It can be seen that the quality of the weld at the weld location can be determined appropriately.

  Next, a specific method for determining the welding quality using the Mahalanobis distance based on the temporal change in the interelectrode voltage measured above will be described with reference to FIG. FIG. 15 is an interelectrode voltage waveform showing temporal changes in the interelectrode voltage at a certain point.

  First, p straight lines parallel to the time axis are drawn at equal intervals with respect to the voltage waveform between the electrodes. Next, the number of intersections at which each straight line intersects the interelectrode voltage waveform is obtained and used as differential data. In addition, the sum of the data amount of the voltage between the electrodes in the area of each straight line or more is obtained as integrated data.

Next, the Mahalanobis distance in the welding is obtained based on the obtained differential data and integral data. Then, the welding quality is calculated from the relationship with the Mahalanobis distance at the spot (for example, 51 spot) for which the welding quality is determined to be good in advance. Specifically, the average value μ and the standard deviation σ are obtained from all the data of the Mahalanobis distance at 51 hit points, and each Mahalanobis distance obtained as described above is assigned with μ + 3σ as a threshold value. Then, as shown in the following equation, the ratio of the Mahalanobis distance data amount (number) Nu included in less than the threshold to the total data amount (number) Na of the Mahalanobis distance at 51 strokes is determined as the welding quality.
Weld quality (%) = (Nu / Na) × 100

  FIG. 16 shows the Mahalanobis distance obtained at the 51th spot (welding quality: good) and the 451th spot (welding quality: poor), as in FIG. In FIG. 16, the Mahalanobis distance was obtained by drawing 1000 straight lines with respect to the voltage waveform between the electrodes at each dot. As shown in FIG. 16, when the temporal change in the voltage between the electrodes is used, a large difference is recognized between the 51st spot in the upper stage (1) and the second (451) spot in the lower stage, unlike FIG. . From this result, it can be seen that by using the Mahalanobis distance based on the temporal change of the interelectrode voltage, it is possible to appropriately determine whether the welding quality is good or bad.

  FIG. 17 shows the result of calculating the welding quality at each spot based on the Mahalanobis distance using the temporal change in the interelectrode voltage. When comparing FIG. 17 and FIG. 10, the weld quality decreases to near 60% with respect to the decrease in the fracture diameter at the 451th hit point, while the weld quality increases with respect to the increase in the fracture diameter after the 851th hit point. It has increased to about 80-90%, and it turns out that the change of a fracture condition and the change of welding quality respond | correspond well. In addition, at the 1351st and 1551th striking points where no nugget was formed, the welding quality was reduced to around 50%, and it can be seen that the determination result represents the difference in nugget formation.

  As described above, by obtaining the Mahalanobis distance using the temporal change in the voltage between the electrodes, it is possible to determine the electrode wear and nugget formation status even in spot welding, and appropriately determine the welding quality. Can do.

  The present invention greatly improves the quality control of welding, which greatly improves the accuracy, efficiency and cost of welding, and thus has a very large industrial applicability.

1 DC stationary spot welder 2 Toroidal coil 3 Weld monitor 4 PC
DESCRIPTION OF SYMBOLS 10 Physical quantity measurement part 20 Physical quantity digitization part 30 Sampling data one-dimensionalization part and vector expression part 40 Quality determination part 80 Electrode 90 Base material 95, 96, 97 The spot which melts by welding 98 Spot weld part 99 Cavity

Claims (7)

A physical quantity measuring unit for measuring a physical quantity related to the welding energy of the welding part during resistance welding;
A physical quantity digitizing section that converts the physical quantity measured by the physical quantity measuring section into sampling data digitized in real time by a predetermined procedure;
A sampling data one-dimensionalization unit that converts the digitized sampling data into one-dimensional data in real time in a predetermined procedure according to the welding conditions;
A vector expression unit for expressing the sampling data converted into one-dimensional data by the sampling data one-dimensionalization unit as a one-dimensional sampling vector in real time according to a predetermined procedure;
Obtaining the magnitude of the difference between the one-dimensional sampling vector expressed by the vector expression unit and the vector serving as the evaluation criterion, and based on the ratio of the magnitude of the difference with respect to the standard deviation of the vector serving as the evaluation criterion of has a determining quality determination unit quality,
The quality judgment unit
When the one-dimensional sampling vector is Rt, the evaluation standard vector is Rs, and the standard deviation is Sp,
Each of Rt, Rs, and Sp is divided into n in time series from 1 to n,
The quality of the weld is determined by comparing the total sum D of D _n obtained by the following equation with a predetermined allowable reference value.
A real-time welding quality judgment device characterized by that.
D _n = | (Rt _n −Rs _n ) | / Sp _n
(However, Rt _n : n-th data of the one-dimensional sampling vector Rt
Rs _n : n-th data of vector Rs serving as the evaluation criterion
Sp _n : nth data of standard deviation Sp) The quality judgment unit
The n-th data Rt _n of the one-dimensional sampling vector Rt is subjected to a predetermined classification according to the elapsed time from the start of welding,
By the total sum D of the weighted D _n is obtained by relative D _n multiplied by the weighting coefficients corresponding to the classification, compared with the allowable reference value is determined in advance, determining the quality of the welding The real-time welding quality determination apparatus according to claim 1 , wherein: A new electrode recognition unit that recognizes that the electrode of the resistance welder has been replaced,
Upon receiving notification that the electrode has been newly exchanged from the new electrode recognition unit, the physical quantity measurement unit, the physical quantity digitization unit, the sampling data one-dimensionalization unit, and the vector expression unit are processed in accordance with the notification. Creates a one-dimensional sampling vector for each predetermined number of welds from the current and voltage at the predetermined number of times after the welding, and then creates a vector as an evaluation criterion based on the created one-dimensional sampling vectors. An evaluation reference vector creation unit for
The quality judgment unit
Wherein upon determination of the quality of the predetermined number of times after the welding, the real-time weld quality determination apparatus according to claim 1 or claim 2, characterized by using a vector in which the evaluation reference vector creation part becomes evaluators created. The physical quantity measuring unit measures voltage and current,
The physical quantity digitizing unit uses the measured voltage and current as sampling data digitized in real time for each welding,
Said sampling data one-dimensional part, claims 1 to 3, the two kinds of sampling data for the digitized voltage and current, characterized in that the one-dimensional data regarding the resistance in real time with a predetermined procedure The real-time welding quality determination apparatus according to any one of the above. A physical quantity measuring unit for measuring a temporal change in the voltage between the electrodes of the welded spot as a physical quantity during spot welding;
A physical quantity digitizing unit that converts the temporal change of the interelectrode voltage measured by the physical quantity measuring unit into sampling data digitized in real time by a predetermined procedure;
Using the digitized sampling data, a Mahalanobis distance calculation unit that calculates the Mahalanobis distance in real time in a predetermined procedure;
Based on the control limit shown in the following formula created from Mahalanobis distance data in spot welding, which is a separately calculated evaluation standard, as a threshold, the Mahalanobis distance data calculated by the Mahalanobis distance calculation unit is sorted and included below the threshold based on the data amount of the Mahalanobis distances, it has a determining quality determination unit the quality of the weld,
The physical quantity digitizing unit uses the temporal change of the interelectrode voltage measured by the physical quantity measuring unit as two-dimensional sampling data of the interelectrode voltage and time digitized in real time by a predetermined procedure,
The Mahalanobis distance calculation unit draws straight lines parallel to the time axis at equal intervals with respect to the two-dimensional sampling data of the interelectrode voltage and time digitized by the physical quantity digitizing unit, The number of intersections with the straight line is obtained as differential data, and the sum of the data amount of the voltage between the electrodes in the area above each straight line is obtained as integral data, and the differential data and the integral data are used in real time. A real-time welding quality judgment device characterized by calculating Mahalanobis distance .
μ + 3σ
μ: Average value of Mahalanobis distance data in evaluation spot welding σ: Standard deviation of Mahalanobis distance data in evaluation spot welding The quality determining unit, according to claim 5, wherein the data amount Nu of the Mahalanobis distance included below a threshold, and based on the total amount of data Na, performing the quality determination seeking weld quality according to the following equation Real-time welding quality judgment device.
Weld quality (%) = (Nu / Na) × 100 A real-time welding quality judgment method, wherein the quality of welding is judged in real time using the real-time welding quality judgment device according to any one of claims 1 to 6 .

Images